UBB pseudogene 4 encodes functional ubiquitin variants

Abstract

Pseudogenes are mutated copies of protein-coding genes that cannot be translated into proteins, but a small subset of pseudogenes has been detected at the protein level. Although ubiquitin pseudogenes represent one of the most abundant pseudogene families in many organisms, little is known about their expression and signaling potential. By re-analyzing public RNA-sequencing and proteomics datasets, we here provide evidence for the expression of several ubiquitin pseudogenes including UBB pseudogene 4 (UBBP4), which encodes Ub^KEKS (Q2K, K33E, Q49K, N60S). The functional consequences of Ub^KEKS conjugation appear to differ from canonical ubiquitylation. Quantitative proteomics shows that Ub^KEKS modifies specific proteins including lamins. Knockout of UBBP4 results in slower cell division, and accumulation of lamin A within the nucleolus. Our work suggests that a subset of proteins reported as ubiquitin targets may instead be modified by ubiquitin variants that are the products of wrongly annotated pseudogenes and induce different functional effects.

Fig. 1. UBB pseudogenes are expressed and encodes ubiquitin variants.

a Expression levels analysis of UBB and UBB pseudogenes 1–5 through RNASeq data available through GTEx Portal from different tissues. b Schematic of UBB and UBB pseudogenes 1–5 mRNA and proteins potentially resulting from translation. Only UBBP3 and UBBP4 encode Ub variants with a diglycine at the C-terminal. The UBBP4 gene contains two ORFs. The first one encodes three Ub repeats, with the third Ub containing a frameshift resulting in a premature stop codon. The second ORF encodes a Ub monomer.

Fig. 2. UBBP4 encodes a functional ubiquitin variant without proteasomal targeting.

a The UBBP4 gene encodes three ubiquitin repeats within a first ORF, Ubbp4^A1, Ubbp4^A2, and Ubbp4^A3, and a fourth ubiquitin within a second ORF, Ubbp4^B1. The differences in amino acids compared with Ub are highlighted in red. b Identification of unique peptides in different large-scale proteomics experiments. The white bars in the peptides indicates the approximate position of the amino acids differences with the canonical Ub, and are shown in red. The unique peptides identified are shown with the reference to the datasets. c Total cell extracts of HeLa cells transfected with HA-tagged Ub, or Ubbp4^A1, Ubbp4^A2, Ubbp4^A3, or Ubbp4^B1 were separated by SDS-PAGE and revealed with an HA antibody (n = 3 biologically independent experiments). d HeLa cells were either untransfected, or transfected with plasmids expressing HA-tagged Ub or Ubbp4^A2 or Ubbp4^B1, and treated or not with MG132. Total cell extracts were separated by SDS-PAGE, and proteins were revealed by immunoblotting with an HA antibody (n = 3 biologically independent experiments). e HeLa cells transfected with either HA-Ub or HA-Ubbp4^B1 were treated or not with MG132 overnight and labeled for immunofluorescence microscopy with an HA antibody. The nuclei were stained with DAPI (n = 3 biologically independent experiments). Scale bars indicate 10 µm. Source data are provided as a Source Data file.

Fig. 3. Ub^KEKS targets proteins with a different specificity compared with Ub.

a HeLa cells were transfected with empty vector (pcDNA), HA-Ub or HA-Ub^KEKS. Proteins were extracted and analyzed by western blot with antibodies against the HA tag or actin (n = 3 independent experiments). b Proteins from SILAC-labeled cells either with light amino acids (control) or with heavier isotopes (M or L) for quantification of proteins enriched in HA-Ub IPs versus HA-Ub^KEKS IPs. c Z-scores of H/L and M/L ratios for each protein detected in all conditions (heavy, medium, and light). Proteins significant (above the 90% percentile) for HA-Ub are indicated in blue, in red for HA-Ub^KEKS, and in green for both. Significance lines are indicated as a blue dotted line (90% percentile for M/L z-scores) and as a red dotted line (90% percentile for H/L ratios). Proteins with a significantly high H/M ratio (specific to Ub^KEKS versus Ub) are indicated as red triangles. Axis are limited to area of interest, thus some values do not appear on this graph. These experiments were performed in biological triplicates (n = 3). d Proteins identified in at least two replicates with a significant ratio of enrichment are represented using a dot plot. The color and size of the dots are proportional to their z-scores and relative abundance, respectively. Source data are provided as a Source Data file.

Fig. 4. Ub^KEKS is important for cell growth and regulates lamin A localization.

a–b Cells were transfected with either GFP-tagged lamin A (LMNA) or lamin B2 (LMNB2), and with HA-tagged Ub or Ub^KEKS. Total cell lysates (Extract) or IPs with GFP-Trap were loaded on SDS-PAGE and revealed with GFP or HA antibodies (n = 3 independent experiments). c Knockout of UBBP4 was performed using CRISPR/Cas9 and two different combinations of guide RNAs in HeLa cells. d Cells are sorted from the debris according to FSC-A and SSC-A, and then single cells are selected using FSC-H and FSC-W for measuring using the FITC channel. Wild type and KO HeLa cells (clones 2.7 and 4.3) were assessed for growth using a CFSE assay at each time points for 96 h. Data are presented as mean values with ±SD. Statistical analysis was performed using two-way ANOVA (F(2,84) = 45.70) and Dunnett’s multiple comparison post hoc tests (n = 8 independent experiments). e Wild type and KO HeLa cells (clones 2.7 and 4.3) were labeled for immunofluorescence microscopy with a lamin A antibody. The nuclei were stained with DAPI (n = 3 independent experiments). f Control and KO HeLa cells (clones 2.7 and 4.3) were labeled for immunofluorescence microscopy with a Nucleolin antibody. The nuclei were stained with DAPI (n = 3 independent experiments). g Quantification of nucleoli was achieved using CellProfiler, through a primary object identification using DAPI and Nucleolin staining, respectively. Each nucleolus was reported to the parental nucleus and the area was determined by measuring the object size shape. Data are presented as box plots where the center lines show the medians; box limits indicate the 25th and 75th percentiles as determined by Prism software; whiskers extend to minimum and maximum values; individual data points are superimposed on the box plot. Statistical analysis was performed using two-way ANOVA (F(2,540) = 37.08) and Dunnett’s multiple comparison post hoc tests (n = 90 cells examined over three independent experiments). Source data are provided as a Source Data file.

Fig. 5. Quantification of Ub and Ub^KEKS in lamin A and lamin B2 pulldown assay.

a Precursors and their fragment ions used for quantification of Ub and Ub^KEKS. b Fragment ion chromatogram of Ub and Ub^KEKS in WT and Ub^KEKS KO (clone 4.3) HeLa cells in lamin A (LMNA) and lamin B2 (LMNB2) pulldown assays. Blue line represents heavy (AQUA) peptide, red line represents endogenous peptides with m/z values indicated in the box above chromatograms. Dotted lines indicate peak boundaries with black arrows showing peaks. Peak values with mass error are indicated above each peak. c Ratio of Ub^KEKS to Ub in WT and Ub^KEKS KO (clone 4.3) HeLa cells in LMNA and LMNB2 pulldown assays. Data are presented as mean values ± SD; individual data points are superimposed on the bar graph. Statistical analysis was performed using unpaired two-tailed t-test (n = 4 independent experiments). Source data are provided as a Source Data file.